Search

Objective: Currently available tumor markers for ovarian cancer are still inadequate in both sensitivity and specificity to be used for population-based screening. Artificial neural network (ANN) as a modeling tool has demonstrated its ability to assimilate information from multiple sources and to detect subtle and complex patterns. In this paper, an ANN model was evaluated for its performance in detecting early stage epithelial ovarian cancer using multiple serum markers., Methods: Serum specimens collected at four institutions in the US, The Netherlands and the United Kingdom were analyzed for CA 125II, CA 72-4, CA 15-3 and macrophage colony stimulating factor (M-CSF). The four tumor marker values were then used as inputs to an ANN derived using a training set from 100 apparently healthy women, 45 women with benign conditions arising from the ovary and 55 invasive epithelial ovarian cancer patients (including 27 stage I/II cases). A separate validation set from 27 apparently healthy women, 56 women with benign conditions and 35 women with various types of malignant pelvic masses was used to monitor the ANN's performance during training. An independent test data set from 98 apparently healthy women and 52 early stage epithelial ovarian cancer patients (38 stage I and 4 stage II invasive cases and 10 stage I borderline ovarian tumor cases) was used to evaluate the ANN., Results: ROC analysis confirmed the overall superiority of the ANN-derived composite index over CA 125II alone (p=0.0333). At a fixed specificity of 98%, the sensitivities for ANN and CA 125II alone were 71% (37/52) and 46% (24/52) (p=0.047), respectively, for detecting early stage epithelial ovarian cancer, and 71% (30/42) and 43% (18/42) (p=0.040), respectively, for detecting invasive early stage epithelial ovarian cancer., Conclusions: The combined use of multiple tumor markers through an ANN improves the overall accuracy to discern healthy women from patients with early stage ovarian cancer. Analysis of multiple markers with an ANN may be a better choice than the use of CA 125II alone in a two-step approach for population screening in which a secondary test such as ultrasound is used to keep the overall specificity at an acceptable level.

Early detection remains the most promising approach to improve long-term survival of patients with ovarian cancer. In a five-center case-control study, serum proteomic expressions were analyzed on 153 patients with invasive epithelial ovarian cancer, 42 with other ovarian cancers, 166 with benign pelvic masses, and 142 healthy women. Data from patients with early stage ovarian cancer and healthy women at two centers were analyzed independently and the results cross-validated to discover potential biomarkers. The results were validated using the samples from two of the remaining centers. After protein identification, biomarkers for which an immunoassay was available were tested on samples from the fifth center, which included 41 healthy women, 41 patients with ovarian cancer, and 20 each with breast, colon, and prostate cancers. Three biomarkers were identified as follows: (a) apolipoprotein A1 (down-regulated in cancer); (b) a truncated form of transthyretin (down-regulated); and (c) a cleavage fragment of inter-alpha-trypsin inhibitor heavy chain H4 (up-regulated). In independent validation to detect early stage invasive epithelial ovarian cancer from healthy controls, the sensitivity of a multivariate model combining the three biomarkers and CA125 [74% (95% CI, 52-90%)] was higher than that of CA125 alone [65% (95% CI, 43-84%)] at a matched specificity of 97% (95% CI, 89-100%). When compared at a fixed sensitivity of 83% (95% CI, 61-95%), the specificity of the model [94% (95% CI, 85-98%)] was significantly better than that of CA125 alone [52% (95% CI, 39-65%)]. These biomarkers demonstrated the potential to improve the detection of early stage ovarian cancer.

A heavy admixture of blood in cervical smears can be problematic for the screener, as the presence of blood can influence the staining quality of the cancer cell nuclei. However, it might also be a blessing in disguise. A retrospective study of 40 clinically important smears, 34 originally signed out as negative for squamous cell carcinoma of the cervix and 6 smears as unsatisfactory, was carried out in comparison with 100 smears from healthy women. Sample parameters were analyzed by macroscopy and neural network scanning. Differences between the two study groups were measured by Pearson's chi(2) test. Of the 40 study cases, one case featured insufficient material, while 16 cases (40%) could confidently be classified as malignant or negative for malignancy. The most important macroscopic parameter of the smears was an admixture of blood. This background feature was also highlighted by the NNS system. Angiogenesis was visualized by the expression of CD34 in many sampled capillary fragments included in the smears. In conclusion, blood in cervical smears may have clinical and diagnostic significance. The rate of "failed smears" in routine cervical screening might thus by CD34 be considerably decreased., (Copyright 2003 Wiley-Liss, Inc.)

The expression of Fas and FasL was studied in 86 patients with benign, borderline, and malignant serous ovarian lesions. Four normal ovaries, and monolayer epithelial cultures from a human fetal ovary, a borderline, and a serous adenocarcinoma were used for comparison. Expression of Fas and FasL was studied immunohistochemically and flowcytometrically. Fas was expressed in all 90 lesions; FasL in 57 lesions, including 2 normal ovaries. Fas expression was significantly increased in borderline tumors compared with benign (P = 0.005, t = -2.94) or malignant serous tumors (P = 0.0001, t = 4.15). FasL expression was significantly increased in malignant tumors compared with benign (P = 0.039, t = -2.10) and borderline tumors (P = 0.0016, t = -3.33). Flow cytometry showed a range of Fas expression in short-term cultures isolated from normal, borderline, and malignant ovarian serous tissue; in the few samples studied, FasL was not expressed. Expression in three serous ovarian cell lines was similar. Fas and FasL expression differed throughout the spectrum of ovarian lesions. FasL expression was increased in malignant tumors, and Fas expression was increased in borderline tumors. Changes in Fas/FasL expression in ovarian surface epithelium might play a functional role in the biology of ovarian tumors., (Copyright 2003, Elsevier Science (USA). All rights reserved.)

Aim: A retrospective study was undertaken to investigate how to improve the diagnosis of endocervical adenocarcinoma in screening programs., Material and Methods: The study group consisted of 29 slides of women diagnosed with cancer but who had negative smears. The slides were subdivided in 12 smears taken less than one year before diagnosis by histology and 17 smears taken between one and 10 years prior to diagnosis. A hundred smears of healthy women were used for comparison. All smears were studied macroscopically after which both groups of smears were scanned by the Neural Network Scanner (NNS). Differences between groups were studied for statistical significance using Pearson's Chi-squared test., Findings: The macroscopic parameter of these smears found to be present most frequently was a heavy admixture of blood. The presence of blood (lysed or not) in the smears was equally consistently highlighted by the NNS. Statistical significance of the association of this parameter, with the presence of cancer, was demonstrated., Conclusion: The awareness of blood as a background feature of adenocarcinoma of the cervix will help to select cases needing special attention. These difficult bloody smears, studied by light microscopy and by NNS images can also be selected for additional MiB-1 staining. With this approach, blood in smears, otherwise frequently leading to a compromise of classification, can become a blessing in disguise. The diagnosis of endocervical adenocarcinoma in screening smears will therefore be improved.

Objective: To study whether interleukin-6 (IL-6) changes the expression of the anti-apoptic Bcl-2 protein in the prevention of cis-diaminedichloroplatinum (II) (CDDP)-induced apoptosis of human ovarian cancer cells., Methods: Comparative studies were performed on 3 ovarian cancer cell lines after 48 hours of exposure to 0.5-50 ng/ml IL-6, 2 micrograms/ml anti-IL-6 monoclonal antibody (anti-IL-6 mAb), 10 microM CDDP, 10 microM CDDP + 0.5-50 ng/ml IL-6, and 10 microM CDDP + 2 micrograms/ml anti-IL-6 mAb. Apoptosis was measured morphologically and by a DNA fragmentation assay. Bcl-2 protein levels were measured by an ELISA., Results: An increase in apoptosis was observed for each cell line after 48 hours of exposure to 10 microM CDDP. Although high doses of IL-6 decreased the percentage of apoptotic cells, this cytokine did not change the expression of the Bcl-2 protein., Conclusion: CDDP-induced apoptosis was negatively controlled by IL-6. However, the anti-apoptic Bcl-2 protein level was not changed by IL-6 in the process of apoptosis in the ovarian cancer cell lines.

A new cell line, FR-car, has been established from a biopsy of a low-grade human cervical squamous intraepithelial lesion (SIL). We confirmed the epithelial origin of the cells by keratin staining using polykeratin, AE1/AE3 and CAM 5.2 antibodies. Sixty percent to 80% of the cultured cells stained positive for proliferative cell nuclear antigen (PCNA) and Ki-67. There was no overexpression of p53. Karyotyping revealed that the cell line was hypodiploid with clonal abnormalities on chromosome 6 and 16. Sections of a biopsy adjacent to the lesion from which the culture was initiated tested positive for human papillomavirus (HPV) 18 DNA by the polymerase chain reaction, but cultured cells tested at several passages were HPV-negative by either type-specific or consensus PCRs. This HPV-negative SIL line may be useful in studies into the cell biology of dysplastic epithelium.

Background: Tumors of borderline malignancy are still a controversial subgroup of ovarian neoplasms. The expression of several cell regulatory proteins was studied to characterize the molecular phenotype of these tumors, and to compare them with their benign and malignant counterparts., Methods: Specimens from 22 patients with tumors of borderline malignancy (11 serous and 11 mucinous tumors), 12 patients with benign tumors, and 16 patients with invasive ovarian carcinomas were evaluated for expression of epidermal growth factor receptor (EGFR), HER-2/neu, PTP1B, and p53 by immunohistochemical techniques., Results: One or both of the tyrosine kinase growth factor receptors EGFR and HER-2/neu was expressed by 42% of benign, 59% of borderline, and 81% of malignant ovarian tumors. EGFR was expressed in a significantly greater fraction of malignant lesions (69%) than borderline lesions (18%) (P< 0.004). EGFR expression was not observed among the 11 mucinous borderline tumors. HER-2/neu was expressed by 50% of borderline tumors and was not a marker for malignancy. The tyrosine phosphatase PTP1B was expressed by a similar fraction of benign (17%), borderline (27%), and malignant (19%) tumors. The number of cases studied precluded correlation of kinase and phosphatase activity. However, among 12 tumors with PTP1B expression, 9 also expressed EGFR or HER-2/neu. Overexpression of p53 was observed only in malignant serous tumors and was not found in malignant mucinous, borderline, or benign lesions., Conclusions: Either EGFR or HER-2/neu was detected in a majority of borderline cancers. PTP1B was present only in a minority of these cancers. Frankly malignant serous lesions differed from borderline and benign tumors with regard to expression of EGFR and overexpression of p53.

Taxol is important in the treatment of both primary and drug-resistant ovarian cancer. Although Taxol is known to stabilize microtubules and block cell mitosis, the effectiveness of this drug exceeds that of other antimitotic agents, suggesting it may have an additional mode of action. Stimulated by murine macrophage studies indicating cytokine induction by Taxol, we have investigated proinflammatory cytokine expression in a series of cell lines and recent explants of human ovarian cancer. Taxol induced secretion of interleukin (IL) 8 but not IL-6, IL-1alpha, or IL-1beta in 4 of 10 samples. Induction was dependent on transcriptional activation, and, in contrast to murine macrophage studies, was apparently independent of an active lipopolysaccharide signaling pathway. Confluent cultures secreted as much IL-8 as proliferating cells. Taxol did not induce IL-8 in breast carcinoma, endometrial stromal, or T-lymphocyte or monocyte cultures. We propose that the local expression of this chemokine in vivo may elicit a host response similar in effectiveness to that of cytokine gene therapy. These data are the first to suggest that a chemotherapeutic agent may have a direct effect on transcription of cytokine and/or growth factor genes in ovarian cancer, and that this effect may not be restricted to proliferating tumor cells.

Magnetic resonance spectroscopy (MRS) can identify abnormal lipoproteins in the plasma of patients with premalignant and malignant tumours. Proteolipid complexes, 8-11 nm and 25-28 nm in size, were isolated from the plasma of a patient with a borderline ovarian tumour. These complexes, which generated a characteristically long MRS T2 relaxation value (greater than 400 ms), were disrupted by ribonuclease. None of the conventional lipoproteins had a T2 value above 160 ms. Chemical analysis of the proteolipid complexes showed a 20% glycolipid component, and MRS identified a fucosylated molecule as the origin of the long T2 value. 9 months after resection of all tumour, a visible lipoprotein band, possibly lipoprotein (a), persisted in the plasma but neither the long T2 relaxation value nor the 8-11 nm or 25-28 nm particles were present. The long T2 relaxation value in the MRS profile, found in isolated proteolipid and unfractionated plasma and serum of other patients with carcinoma of the ovary and colon, provides a non-invasive method of assaying for cancer.

A cell line derived from a human ovarian carcinosarcoma was established in tissue culture and in nude mice. Two sublines, LDF and HDF, separated by discontinuous density centrifugation were also established from the parent line JoN. The cloning efficiency of the JoN line was 21%. Morphologic features of adenocarcinoma cells characteristic of the parent JoN cells were retained in the sublines and clones; all lines showed the same karyotype and DNA content (pseudodiploid and pseudotetraploid). Keratin, as demonstrated immunohistochemically, was strongly expressed in the parent line JoN and the xenograft tumor, but not at all in the LDF sublines and only moderately in the HDF sublines. Vimentin, however, was expressed in neither the parent line JoN nor the xenograft tumor, but was present in both sublines. Transglutaminase and plasminogen activator activity was high in the parent line JoN. Neither, sublines nor clones showed the same high enzyme activity as the parent line. It is concluded that this human tumor line JoN is comprised of epithelial cells, capable of multidirectional differentiation.

Six cell lines, established from four primary ovarian carcinomas were examined cytogenetically. The lines varied greatly in their chromosome complement. All cells from the lines were aneuploid, although one cell line contained two populations having a pseudodiploid and a pseudotetraploid modal chromosome number. Every chromosome group was involved with loss and gain of chromosomes, but some individual chromosomes were more prone to aneuploidy than others. Chromosome #6 was the most stable throughout. Structural changes gave rise to many marker chromosomes. Although most markers were random and the majority unidentifiable, some abnormalities of clonal origin were found. Deletions especially of chromosome #1, were the most common change. Further sequential studies may elicit the origin, stability, and timing of the chromosome abnormalities.

The first xenograft line of small cell undifferentiated carcinoma of the prostate (UCRU-PR-2) has been established and characterized. The donor tumor and the xenograft share the common morphological and ultrastructural features of small cell undifferentiated carcinoma (including neurosecretory granules) but also elaborate epithelial membrane antigen and carcinoembryonic antigen, in addition to neurone-specific enolase. The line expresses a diploid DNA complement. Androgen and estrogen receptors are not expressed, although prostatic acid phosphatase is present in sera from tumor-bearing mice in low levels. From these studies, we postulate a possible common stem cell origin for adenocarcinoma and small cell undifferentiated carcinoma of the prostate; further studies of a cell line derived from this tumor may clarify the issue.

Cytogenetic analysis of a human ovarian carcinoma cell line JoN was performed at passages 6, 43, and 89. At passage 6 there were pseudodiploid and pseudotetraploid cells containing up to 22 markers. Pseudodiploid cells predominated. Significant differences were seen in the later analyses. At passages 43 and 89, cells had a modal number of 44 and 88 chromosomes with four new reciprocal translocations and loss of all unidentified markers. There was retention of one identified marker throughout plus one that gradually disappeared. Karyotypes at passages 43 and 89 were identical, suggesting stability of rearrangements. These changes indicate continuing chromosomal rearrangements in culture, not necessarily creating more complexity.

Human ovarian cancers of four different histological types have been cultured in vitro and in nude mice. Nineteen tumor specimens (11 solid tumors and eight malignant effusions) were obtained from 14 patients. Tumor lines from ten of these patients were established after several subpassages, and six xenograft lines have been grown, all of them from tumors of which a cell line exists in vitro. In all, 14 lines have been successfully cultured from the 19 tumor specimens. The morphology (studied by light and electron microscopy) of the established lines in vitro and in vivo resembles that of the original tumors in all cases. Flow cytometric studies of DNA content of the original tumor specimens and the cell culture and xenograft lines were performed. In all lines in both culture systems, aneuploid cells became predominant after the first to fourth passages, despite an aneuploid peak having been evident in only nine of the 19 initial specimens. Four of the original tumor specimens contained measurable estrogen receptors and five progesterone receptors, but none of the established cell lines expressed these hormone receptors. These results indicate that, while morphological features are similar in the initial tumor specimens and in the established lines in vitro and in vivo, flow cytometric and steroid hormone receptor data suggest selection of aneuploid and receptor-negative cells.